Potential of lower-limb muscles to accelerate the body during cerebral palsy gait

Abstract

Two of the most common gait patterns in children with spastic diplegic cerebral palsy (CP) are termed ‘crouch gait’ and ‘jump gait’. While outcomes of surgical interventions designed to improve functional mobility are generally positive, many children displaying these gait patterns show minimal or no improvement post-surgery. A poor response to treatment may be partially attributable to incorrect interpretations of muscle function. Computational techniques that assess muscle function may help address this issue, but before studying specific surgeries, the gait patterns themselves must be better understood. The aim of this study was to identify differences in lower-limb muscle function when comparing crouch, jump and able-bodied gait patterns by quantifying the potential of lower-limb muscles to accelerate the body's center of mass. A muscle's potential acceleration was defined as the acceleration induced by a unit of muscle force. Dynamic simulations of walking using musculoskeletal models were developed for eight children with crouch gait, ten with jump gait, and ten controls. There were significant differences (p < 0.05) in muscle potential accelerations between crouch and able-bodied gait patterns, and between jump and able-bodied gait patterns, for most of the major muscles of the hip, knee, and ankle. One important outcome was the identification of the significantly reduced potential of gluteus medius to extend the hip in both crouch gait and jump gait. Potential acceleration analyses appear to be suitable for evaluating differences between common gait patterns and may also be applied to study the effects of surgical treatments. The results of such studies may lead to improved treatment outcomes for individuals with impaired mobility.

Introduction

Gait patterns in children with spastic diplegic cerebral palsy (CP) have been classified in many different ways for several decades [1], [2]. Two of the most common patterns in the various classifications are termed ‘crouch gait’ and ‘jump gait’. These two gait patterns may be differentiated on the basis of stance phase sagittal-plane kinematics [3]. Specifically, although both patterns are characterized by excessive hip and knee flexion, in jump gait there is excessive plantarflexion at the ankle, whilst in crouch gait there is excessive dorsiflexion at the ankle [3]. Pharmacological and/or surgical interventions may be prescribed to children displaying these gait abnormalities [4]. Treatment decisions are often based on functional assessments of the neuromusculoskeletal system such as quantitative gait analysis. While quantitative gait analysis can be used to measure lower-limb joint kinematics, moments and muscle electromyographic activity [5], the function of individual lower-limb muscles during a task such as walking can only be inferred from these data. The outcomes of surgical interventions for crouch gait are generally positive in terms of a correction toward a more extended posture at the hip and knee, with improvements in knee pain and increased functional mobility [6], [7]. However, the outcomes of surgery for some children with crouch gait results in minimal or no improvements, which may be partially attributable to incorrect interpretations of lower-limb muscle function (e.g., scenarios described by Arnold [8] and Delp et al. [9]).

Musculoskeletal modeling has the potential to address this issue by providing non-invasive quantification of muscle function [10]. Person-specific models combined with quantitative gait analysis data enable the development of computer-based simulations of motion. Muscle forces can be determined using a variety of optimization-based techniques [11], and neuromuscular coordination patterns can be revealed through muscle contributions to the acceleration of the body's center of mass (e.g., [12], [13]). For example, Liu et al. [14] used this technique to determine the muscles that provide deceleration and/or forward propulsion during normal gait, and Steele et al. [15] similarly determined the muscles that are important in crouch gait. Unfortunately, as children with CP display neural function that is clearly sub-optimal, the ability of optimization-based techniques to accurately represent lower-limb muscle function in this population is limited. One way to overcome this limitation is to avoid mathematical optimization, and instead evaluate the potential or capacity of individual lower-limb muscles to accelerate the body.

The potential accelerations of the body's center of mass induced by individual muscles describe the influence of musculoskeletal geometry and limb positions on the functional capacities of these muscles, as distinct from the neuromuscular coordination patterns revealed by muscle-force-induced accelerations. To the best of our knowledge, the only investigation of gait patterns using potential center-of-mass accelerations was performed by Lakin et al. [16], despite several studies having evaluated hip and knee accelerations (e.g., [17], [18]). While Lakin et al. [16] reported large differences between crouch gait and normal gait, only the sagittal-plane accelerations induced by three lower-limb muscles were evaluated. A logical extension of this work is a more complete comparison of not only crouch gait and normal gait, but jump gait as well. It is anticipated that outcomes may provide novel clinical insights, and would generate baseline data for future work targeted at specific clinical questions, such as the effect of a particular surgical procedure.

The aim of this study was to determine the differences in lower-limb muscle function between jump gait and crouch gait by quantifying the potential of key lower-limb muscles to accelerate the body's center of mass. Our hypotheses were: firstly, that there would be significant differences between the two cohorts with abnormal gait patterns when compared with able-bodied counterparts, particularly for those muscles that typically require corrective treatment (e.g., rectus femoris and hamstrings); and secondly, that the potential accelerations in crouch and jump gait would differ most significantly for the ankle-spanning muscles, because ankle position is the main differentiating factor between these patterns.